1 PROJECT ABSTRACT 23 disease and related dementias (ADRDs). Nevertheless, almost nothing is known about the impact of social 4 integration on the risk of ADRDs for older Mexican-origin adults. Latinos, including Mexican-origin individuals, Greater integration within family and community networks has been linked to lower risk of Alzheimer's 5 will account for nearly 20% of the older adult population by 2050. There is consequently a critical need to 6 understand the impact of social factors on ADRDs risk for this increasingly important segment of the U.S. older 7 adult population. This research should take into account the specific context of social networks for Mexican- 8 origin older adults. In particular, many Mexican-origin older adults in both the U.S. and Mexico may have 9 geographically dispersed family networks as the result of U.S. migration. Family members living across borders 10 may continue to serve as sources of emotional support. On the other hand, dispersed family networks may 11 entail diminished emotional supports for older adults. Surprisingly little is known about how social integration 12 within local or cross-border family networks impacts cognitive outcomes. The research component of this 13 mentored career development award will apply novel epidemiologic tools to quantify the impact of local and 14 cross-border social integration on cognitive outcomes for older Mexican-origin adults in the U.S. and Mexico 15 and identify the degree to which these effects may be explained by or modified by late-life depression. The 16 proposed research will also examine the potential for both local and cross-border social integration to buffer 17 the adverse impact of early-life socio-economic disadvantage (ELSED) on cognitive outcomes for Mexican- 18 origin older adults. This mentored career development award proposes three areas of training that will allow 19 the candidate to develop an independent research career focused on the epidemiology of cognitive aging in 20 cross-national contexts: 1) cognitive aging subject matter, 2) causal inference and causal mediation methods, 21 and 3) cross-national data harmonization. The specific aims will be achieved using data from two longitudinal 22 studies of older Mexican-origin adults in the U.S. and Mexico ? the Sacramento Area Latino Study on Aging 23 and the Mexican Health and Aging Study ? that are unique in their inclusion of both local and cross-border 24 social integration measures. These studies will be used individually and as part of a harmonized cross-national 25 dataset to carry out the following aims: 1. Quantify the effects of local and cross-border social integration on 26 cognitive function and cognitive impairment. 2. Estimate the degree to which the effects of local and cross- 27 border social integration on cognitive function are a) explained by and b) modified by depression. 3. Test the 28 capacity for local and cross-border social integration to modify the impact of ELSED on cognitive function. The 29 expected outcome is a rigorous cross-national study of the longitudinal impact of social integration on cognitive 30 outcomes for Mexican-origin older adults. Aim 3 will also serve as a methodological test case for research on 31 modifiable factors that might buffer the adverse impact of ELSED on cognitive aging.